Distance to subject in E5's EXIF

[katran]

Sometimes it is very interesting to know wich was the distance to the subject from a photo.
Some cameras (Nikon) have this distance recorded in EXIF.

I thought that this information is not available in Olympus cameras, but I was wrong.

Tordan58 discovered that Olympus Viewer removes the distance_to_subject tag from exif when converting raw into jpg.

This tag is saved in JPEG only if the conversion is done with DXO OPTICS PRO program.

Example :
9206 37382 ExifSubjectDist 5 4220/1000

Here the distance to the subject was 4.22 meters.
Maybe Tord can comment more.

Many thanks to Tord for discovering this.

[Tordan58]

Re: Distance to subject in E5's EXIF

Hi,

This EXIF tag is (or seems to be) present in all E- 4/3 cameras RAW files. It is present in E520, E620, E3, E5 and I suspect it is there in the others as well.

For some reason this tag is discarded when developing RAWs in Olympus Viewer 2. Does anyone use other Olympus SW and could check if this tag is preserved in the generated JPG?

It would be interesting to see if the tag is supported by other SW. I know members on this forum are using other SW packages such as Lightroom, Photoshop etc.
- Could you check if the tag is preserved?

The tag identifier is HEX 9206, or if you prefer, decimal 37382. The value of the tag is a text string representing a distance, coded as NNNNN/1000. For instance 4220/1000 for 4.22 meters.

The name of the tag in the JPG is ExifSubjectDist but I guess this name is given by DXO.

It could be that DXO Optics Pro has a unique feature...

/Tord

PS on this subject there is a post I created yesterday, asking about reference documentation to ORF files.

[DerekW]

Re: Distance to subject in E5's EXIF

When viewing an image in Aperture and the EXIF info is displayed there is a field called Focus Distance.

Because of the way that Aperture worls this information is available regardless of how the image has been processed as (I think ) the information is stored in the Aperture database.

The Image Info displays in Aperture can be edited so that EXIF and subject info can be shown on the side panel for each image.

Quite useful.

[birdboy]

[DerekW]

Re: Distance to subject in E5's EXIF

Looking closely at the reported focus distance for two images - one from about 6 feet away and one from about 30 feet away. The reported distance for the distant picture was closer than for the nearer picture. The numbers did have an approximate symbol preceding the number.

[Tordan58]

Re: Distance to subject in E5's EXIF

Hi,

The EXIF reported distance in photos I have taken seems reasonably in line with the true one.

I suppose the only way to find out is to proceed with measurements... or to calculate the distance by knowing the focal length (which can be translated to an angle of view) and knowing, or assessing, the size of elements in the picture.

Anyway, I cannot find an application for the tag how to improve image quality in post-processing

But it
/Tord

[katran]

Re: Distance to subject in E5's EXIF

I noticed that KUSO EXIF VIEWER can also display the FocusDistance from the raw files. This is an small-free program wich will work for anyone.

What is bad, is that all distances do not look real - are much smaller than the real distances.
I have a picture where in exif it is written 11 meters, but in reality there was 40-50 meters.
So either the information from the exif is wrong, either we have to multiply with a factor (maybe 4 ?).

We can do precise measurings an see if we can find such a factor wich gives the corect distance.

[Tordan58]

Re: Distance to subject in E5's EXIF

Hi

Perhaps we should start by comparing different EXIF viewers and see if we get the same numeric values out from them. It could be they are reporting using different measurement units, it could be they are interpreting the data in different ways... Maybe there are linear correlations... Maybe some of them got it wrong...

What we also could do is to analyze some pictures containing pictural elements of known size (or a size possible to approximate) and reverse calculate the distance which is easy if the focal length is known.

/Tord

[birdboy]

Re: Distance to subject in E5's EXIF

I had thought of doing some measurements but my tape measure only goes up to 5m. Its a pretty dull day today and I have no rooms in my home of that lenght.

I think that below 10m they give quite a good guide but not accuracy, above that is guess work. Just look at the divisions on the lens barel for a 50-200SWD lens and the small distance between 10m and infinity. Quite how you interpolate between these two values is beyond me when one value is not specific.

I have taken pictures of Jumbo jets circulating over my home and the exif reports the distance to 50m (no I do not live on the end of Heathrow runway). The pictures I took of the moon shows the distance to be 16m.

My point about Olympus and other manufactures building distance measurement in would require them to do some internal measuring of how much signal was applied to the stepper motor that drives the focus mechanism.

John

[DerekW]

Re: Distance to subject in E5's EXIF

The reported focus distance changes according to whether the lens is focussing down from infinity or up from close up. I found a 1 metre difference in a 2.5, 3.5 metre distance

[birdboy]

Re: Distance to subject in E5's EXIF

The reported focus distance changes according to whether the lens is focussing down from infinity or up from close up. I found a 1 metre difference in a 2.5, 3.5 metre distance

We used to call this hysteresis
John

[DerekW]

Re: Distance to subject in E5's EXIF

so did we - along with the word "losses"

[Imageryone]

Re: Distance to subject in E5's EXIF

Not any kind of techie on this, but I have always used Opanda as my exif viewer and as far as my limited memory works, the distances are aways about correct.